Semen contains cationic amyloid fibrils derived from proteolytic cleavage fragments of prostatic acid phosphatase (PAP) that efficiently capture HIV-1 particles. These fibrils have been termed semen-derived enhancer of viral infection (SEVI) and enhance HIV-1 infectivity by several orders of magnitude. It is expected HIV-1 virions in semen will efficiently form complexes with SEVI, and that the transmitted form of HIV-1 in semen may therefore correspond to virus-SEVI complexes. We hypothesize that virus-SEVI complexes will present novel antigenic structures, different from those present on free virions or conventional recombinant antigens, and that the virus-SEVI complex can be targeted to derive an innovative vaccine capable of protecting against HIV transmission via semen. By targeting the complex between a host protein and the virus, we expect to elicit an immune response that recognizes the authentic form of HIV-1, as it is found in semen, and - potentially - to circumvent the problem of viral antigenic variation. Two aims are proposed. In aim one, we will test whether antibodies can block SEVI-mediated enhancement of HIV-1 infection. To do this, we will first identify PAP residues essential for enhancement of HIV-1 infectivity by SEVI. This will be achieved by testing whether fibrils formed by mutated derivatives of the amyloidogenic PAP peptide can enhance the efficiency of HIV-1 infection. Since other amyloid fibrils (Ass) have adjuvant-like immunostimulatory properties, we will also test whether these mutated fibrils elicit the production of inflammatory cytokines. We will then immunize guinea pigs with SEVI, or with short linear peptides derived from the virus-binding surface of SEVI, and test whether the elicited antibodies prevent SEVImediated enhancement of HIV-1 infectivity and/or SEVI-mediated immune activation. In aim two, we will evaluate whether virus-SEVI complexes, or mimics thereof, can elicit protective, virusneutralizing antibodies. To do this, we will assemble complexes between SEVI and virus particles, and also between SEVI and oligomeric gp140 (as a surrogate for the virus-SEVI complex). We will then evaluate the antigenicity of these complexes using competitive antibody-binding ELISA assays and compare results to those obtained with free gp140 oligomers or virus particles. In Aim 2B, we will immunize guinea pigs with gp140-SEVI (or virus-SEVI) complexes and assess the strength and quality of the elicited humoral immune response (including neutralizing antibodies);results will be compared to animals receiving gp140 or SEVI alone. Finally, in Aim 2C, we will test whether a SEVI-based vaccine is immunogenic in rhesus macaques and whether it can protect against vaginal transmission of SIV. Collectively, these studies will provide a conclusive proof-of-principle test for our proposed approach.